I finally got around to testing. It turns on that not for the first time I was over complicating things. I don't actually need the black hole route unless I want to see a particular remote gateway as down for any particular reason. So my basic configuration is ...
/ip route
add check-gateway=ping comment="*** Default Route - recursive ***" distance=2 gateway=8.8.8.8
add check-gateway=ping comment="*** Default Route - recursive ***" distance=2 gateway=1.1.1.1
add check-gateway=ping comment="*** Default Route - Static ADSL ***" distance=3 gateway=<Address of ADSL router>
add check-gateway=ping comment="*** Remote Gateway ***" distance=1 dst-address=1.1.1.1/32 gateway=<Address of LTE router> scope=10
add check-gateway=ping comment="*** Remote Gateway ***" distance=1 dst-address=8.8.8.8/32 gateway=<Address of LTE router> scope=10
If the LTE router is completely down, the two Remote Gateway routes show as unreachable, so do the two recursive default routes. However from an inside host on the LAN I can still ping 1.1.1.1 or 8.8.8.8. This shows that it is using the default route for those destinations, but this is not messing up the recursive routes.
I might still add the black hole route for at least one of the remote gateways, that way losing contact with that destination would be an alert that the Internet has failed over.